Hacker News Reader

Comment by daemonk | Hacker News Reader

Cell-free synthesis is the next step. Theres slow progress, but very interesting operational models if it works.

I once visited a lab near Berlin where they told me that cell-free synthesis is basically like cell-based synthesis, just that they kill the cells first and use what's inside. The major benefit, I was told, was that you don't have to keep the cells alive.

It sounds very odd at first glance but seems very fascinating to me (and I know way less about biology than I would like to), this idea of dissecting a system into its pieces because we know how to use the pieces in a superior way to the original system itself.

I don't like that description very much. The cell being "alive" is literally the top reason the cellular machineries work so well. By being "alive" it is keeping the environment optimal and perfect for the enzymes to run and being productive. Taking this away and then seeing problems about efficiency is like taking off the wheels and then asking why the car is not running well.

They are trying to fix it by maintaining the same cytoplasmic conditions using all the machines and sensors in a bioreactor. But that can scale is a different question. The narrow ranges of conditions biological processes require is already hard to maintain while the cells itself is still whole and using more than half of its enzymes to do this.

Think of it more as just getting to the one function that you are using. I think a cell free technique that people might have heard of even in high school biology at this point is the polymerase chain reaction. Rather than have some cell we grow in some growth media under certain conditions optimal for cell growth and harvest cloned DNA from these cells, we have isolated the involved protein directly (DNA Polymerase). We know exactly what temperature is most optimal for each step, and infact depending on what targets of DNA we are looking at certain temperatures may be more optimal for these targets. We know the exact pH that these molecules are most effective at and this might not even be a pH found in the cell but one determined empirically by measuring enzyme activity across a gradient of pH and/or temperature ranges among other potential conditions. We know the inputs and reaction and can supply just enough of these molecules as we need to drive satisfactory yield in order to decrease our costs. We can optimize the timing of this temperature-dependent reaction again to the specifics of the DNA molecule that we are amplifying. We can change how quickly this reaction is heated up or cooled down to optimize for product yield. We can change out components of this reaction such as our enzyme for ones that might operate faster with less accuracy, or ones that operate slower with higher fidelity, or ones found from other species that might have these properties, or engineered proteins that might have these properties.

Many reasons why you'd want to go cell free. You are no longer binned into the cell but are now just throwing things into a tube where you can control all the parameters that matter for your desired function.

Biology really is the ultimate modular system!

I have strong doubt it will work. On paper we are just removing an arbitrary boundary that is the cell membrane/wall and combine trillions of cells together to form a giant cell. So it sounds very reasonable and attractive in engineering terms. But these are biological systems inherently designed/evolved to work in one special condition that is the cytoplasm of the host...

Any engineer would know the kind of troubles that can occur if we remove a critical principle of the original design that was there since the system's inception and is literally hardcoded on an operational/abstract level into the system's functions. Trying to fix that mess is unlikely to yield anything close to the original and requires manipulating techniques and knowledge highly intimate within the system.

And we know barely anything about the intimate and detail operation of the molecular structures inside a cell. We know the broad strokes. But ask any biochemist if they are confident of the entire reaction pathway of any enzymes down to the thermodynamics calculations associated with all conformational changes. Maybe you will find a dozen enzymes that we can describe at this detail.

And to optimize cell free synthesis, we would need to know of tens of thousands more. It is possible, but it isn't something I would expect to see soon unless there are major breakthroughs.

All that matters is that it works and is easier or cheaper than existing cell based techniques. If it does, whats the harm then? This is why people are figuring this out on grant funded work after all, to see if it is in fact cheaper than existing techniques. This is how we have found many optimizations to lab work and other processes through basic research in the past.

TXTL systems already works. And there are antibodies (not many) that can only be produced cell free.

The problem is really scale.

Cell-free synthesis is really nice for R/D (it's fast!) but honestly kinda sucks for producing things. Cells have a really nice property of making more of themselves for the cost of basic chemicals (salt, sugar, nitrogen).

TXTL also can be pretty finicky generally - cold storage is required, have to store chemicals separately, ribosomes tend to fall apart and not get recycled, etc. Cells will heal themselves perfectly (sometimes too perfectly, kicking out your construct)

The idea of using a controlled, cell-free environment to churn out proteins or antibodies is so elegant

Reply on news.ycombinator.com